Validation of MRI predictors of multiple sclerosis diagnosis in children with acute CNS demyelination

BACKGROUND: In a recent Canadian prospective study of children with acute demyelinating syndromes (ADS), we demonstrated that the presence of T2 periventricular and T1-hypointense lesions predicted MS diagnosis. We aimed to validate these predictors in a Dutch cohort of children with ADS. METHODS: Participants with ADS were identified from a prospective cohort or archived dataset. MS was diagnosed based on clinical or MRI evidence of relapsing disease. Baseline MRI scans were evaluated for the presence of the two predictive parameters. Sensitivity, specificity, positive (LR+) and negative likelihood ratios (LR-), and positive (PPV) and negative predictive value (NPV) were calculated to evaluate the performance of the MRI parameters at classifying children as having MS or monophasic demyelination. FINDINGS: Of 115 children identified with ADS between December 1993 and December 2009, MRI scans from 87 children (45 prospective; 47 archived) were evaluated; scans of 28 children were excluded due to incomplete or poor quality imaging. Mean duration of observation was longer in the archived group (7.1 years, SD 3.5) than the prospective cohort (3.3 years, SD 1.4). 30 children were diagnosed with MS. Performance of the parameters was not statistically different between the prospective cohort (sensitivity 93.3% [68.1-99.8]; specificity 86.7% [69.3-96.2]; LR+ 7.0 [2.8-17.6]; LR- 0.08 [0.01-0.5]; PPV 77.8% [52.4-93.6]; NPV 96.3% [81.0-99.9]) and archived group (sensitivity 66.7% [38.4-88.2]; specificity 85.2% [66.3-95.8]; LR+ 4.5 [1.7-11.9]; LR- 0.4 [0.2-0.8]; PPV 71.4% [41.9-91.6]; NPV 82.1% [63.1-93.9]). INTERPRETATION: In an independent Dutch cohort, we confirm that the presence of ≥1 T2 periventricular and ≥1 T1-hypointense lesions reliably identifies children with MS. FUNDING: Dutch MS Research Foundation.